Biomechanics of transvertebral screw fixation in the thoracic spine: an in vitro study.

Abstract

OBJECTIVE Transvertebral screws provide stability in thoracic spinal fixation surgeries, with their use mainly limited to patients who require a pedicle screw salvage technique. However, the biomechanical impact of transvertebral screws alone, when they are inserted across 2 vertebral bodies, has not been studied. In this study, the authors assessed the stability offered by a transvertebral screw construct for posterior instrumentation and compared its biomechanical performance to that of standard bilateral pedicle screw and rod (PSR) fixation. METHODS Fourteen fresh human cadaveric thoracic spine segments from T-6 to T-11 were divided into 2 groups with similar ages and bone quality. Group 1 received transvertebral screws across 2 levels without rods and subsequently with interconnecting bilateral rods at 3 levels (T8-10). Group 2 received bilateral PSR fixation and were sequentially tested with interconnecting rods at T7-8 and T9-10, at T8-9, and at T8-10. Flexibility tests were performed on intact and instrumented specimens in both groups. Presurgical and postsurgical O-arm 3D images were obtained to verify screw placement. RESULTS The mean range of motion (ROM) per motion segment with transvertebral screws spanning 2 levels compared with the intact condition was 66% of the mean intact ROM during flexion-extension (p = 0.013), 69% during lateral bending (p = 0.015), and 47% during axial rotation (p < 0.001). The mean ROM per motion segment with PSR spanning 2 levels compared with the intact condition was 38% of the mean intact ROM during flexion-extension (p < 0.001), 57% during lateral bending (p = 0.007), and 27% during axial rotation (p < 0.001). Adding bilateral rods to the 3 levels with transvertebral screws decreased the mean ROM per motion segment to 28% of intact ROM during flexion-extension (p < 0.001), 37% during lateral bending (p < 0.001), and 30% during axial rotation (p < 0.001). The mean ROM per motion segment for PSR spanning 3 levels was 21% of intact ROM during flexion-extension (p < 0.001), 33% during lateral bending (p < 0.001), and 22% during axial rotation (p < 0.001). CONCLUSIONS Biomechanically, fixation with a novel technique in the thoracic spine involving transvertebral screws showed restoration of stability to well within the stability provided by PSR fixation.